Relational Teaching in the Age of AI: A Model for Feedback-Driven Learning

Abstract

In an era where artificial intelligence is rapidly transforming teaching and learning, maintaining meaningful relationships with students remains a core responsibility of educators. This presentation explores how AI-supported assessment tools—specifically Co-Grader and ChatGPT—can be used to enhance, rather than replace, relational teaching in secondary science education. AI is often viewed as depersonalizing, yet when used intentionally, it can create more space for teachers to engage with students as individuals. This project examines my integration of AI to support scientific writing and methodology instruction for secondary students. I employ ChatGPT to generate clear, descriptive assignment prompts, scaffold complex tasks, and create multi-point rubrics aligned with NGSS science and engineering practices. Co-Grader is then used to evaluate student submissions quickly and consistently. The critical relational layer follows: I use ChatGPT to personalize Co-Grader’s standardized feedback, tailoring it to each student’s strengths, areas for growth, and demonstrated learning patterns. This allows me to deliver feedback that feels human, encouraging, and academically rigorous—without compromising time needed for authentic teacher-student interaction. Preliminary outcomes show improved student engagement with feedback, more frequent revision cycles, and stronger performance in scientific reasoning and written communication. Students report that the individualized comments “feel like the teacher sees me,” highlighting the continued importance of relational pedagogy. This presentation will provide examples of AI-generated feedback, discuss ethical considerations, highlight implications for teacher preparation, and offer a framework for relational AI-supported assessment that other educators can adopt.

Ähnliche Arbeiten